
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Steve Tran Blog</title>
      <link>https://nhuantranduc.dev/blog</link>
      <description>All about Data and Artifical Intelligence</description>
      <language>en-us</language>
      <managingEditor>tranducnhuan@hotmail.com (Steve Tran)</managingEditor>
      <webMaster>tranducnhuan@hotmail.com (Steve Tran)</webMaster>
      <lastBuildDate>Sat, 28 Mar 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://nhuantranduc.dev/tags/rust/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://nhuantranduc.dev/blog/x-algorithm-part-2</guid>
    <title>X Algorithm Part 2: The Composable Pipeline Framework in Rust</title>
    <link>https://nhuantranduc.dev/blog/x-algorithm-part-2</link>
    <description>How X built a reusable, trait-based pipeline framework in Rust that separates orchestration from business logic — with clever error handling that prioritizes availability over correctness.</description>
    <pubDate>Sat, 28 Mar 2026 00:00:00 GMT</pubDate>
    <author>tranducnhuan@hotmail.com (Steve Tran)</author>
    <category>x-algorithm</category><category>rust</category><category>system-design</category><category>recommendation-system</category>
  </item>

  <item>
    <guid>https://nhuantranduc.dev/blog/x-algorithm-part-3</guid>
    <title>X Algorithm Part 3: Candidate Sourcing — Thunder and Phoenix Retrieval</title>
    <link>https://nhuantranduc.dev/blog/x-algorithm-part-3</link>
    <description>How X finds ~200 candidate posts from billions in milliseconds — Thunder keeps in-network posts in memory, while Phoenix Retrieval uses a two-tower transformer to discover out-of-network content via embedding similarity search.</description>
    <pubDate>Sun, 29 Mar 2026 00:00:00 GMT</pubDate>
    <author>tranducnhuan@hotmail.com (Steve Tran)</author>
    <category>x-algorithm</category><category>recommendation-system</category><category>machine-learning</category><category>rust</category>
  </item>

  <item>
    <guid>https://nhuantranduc.dev/blog/x-algorithm-part-5</guid>
    <title>X Algorithm Part 5: Scoring, Filtering, and the Final Feed</title>
    <link>https://nhuantranduc.dev/blog/x-algorithm-part-5</link>
    <description>How X transforms raw ML predictions into a polished, personalized feed — the four-scorer chain, 10 pre-scoring filters, and the waterfall that reduces ~200 candidates to ~40 ranked posts.</description>
    <pubDate>Fri, 03 Apr 2026 00:00:00 GMT</pubDate>
    <author>tranducnhuan@hotmail.com (Steve Tran)</author>
    <category>x-algorithm</category><category>recommendation-system</category><category>system-design</category><category>rust</category>
  </item>

    </channel>
  </rss>
